A system for intensity modulated dose plan verification based on an experimental pencil beam kernel obtained by deconvolution

Abstract

The number of intensity modulated radiation therapy ͑IMRT͒ procedures is continuously growing worldwide and it is necessary to develop tools for patient specific quality assurance ͑QA͒ that avoid using machine time that could be employed in treating additional patients. One way of achieving this goal is to perform a multileaf collimator quality assurance periodically in the linear accelerator and check the treatment planning system ͑TPS͒ calculation by employing an independent calculation system. Within the work frame of the pencil beam kernel approach, a new system was developed for obtaining an experimental kernel. This new technique is based on a deconvolution procedure using the Hankel transform. The resulting kernel is obtained in a way completely independent of those employed in commercial treatment planning systems, usually calculated by Monte Carlo simulations. Also provided are comparisons between calculated and measured doses with radiographic film, linear array of diodes, and ionization chamber. Measurements taken in polystyrene and water for clinical IMRT plans demonstrate that this method can calculate IMRT dose distributions, as well as treatment times, with great accuracy. Apart from other applications, it can be used as a double-check algorithm for IMRT QA.

Intensity modulated dose calculation with an improved experimental pencil‐beam kernel

“…The algorithm used here has been extended on the basis of previous work 6 and differs from it in several ways. The absorbed dose is calculated using the following expression:…”

“…6 The term s e includes the characterization of the finite extent of the primary source as well as that of the head scatter source. Although the algorithm provides a function w that is not exactly the kernel k, this approach enables dose calculations for any fluence distribution.…”

“…This approach has certain limitations that have been discussed elsewhere and that motivated the authors to develop a new method for deconvolving the kernel. 6 In the model presented in Ref. 6 and improved here, the pencil-beam kernel was obtained using a deconvolution procedure with dose measurements in a circular field 50 mm in diameter and modeling the incident fluence as a 2D circular step function.…”

“…6 In the model presented in Ref. 6 and improved here, the pencil-beam kernel was obtained using a deconvolution procedure with dose measurements in a circular field 50 mm in diameter and modeling the incident fluence as a 2D circular step function. The Hankel transform was used to obtain the kernel.…”

“…[1][2][3][4][5] They use tissue-phantom ratio values [1][2][3] or a pencil-beam approach, where the kernels have been empirically fitted to mathematical functions 4 or modeled using Monte Carlo methods. 5 In a previous study, 6 the authors presented a pencil-beam approach that enables the calculation of two-dimensional dose distributions. This approach is based on an experimental kernel obtained by deconvolution.…”